Asynchronous Traffic Shaping and Redundancy: Avoiding Unbounded Latencies in In-Car Networks

TL;DR

The paper tackles unbounded end-to-end latencies that arise when pairing Asynchronous Traffic Shaper (ATS) with Frame Replication and Elimination for Reliability (FRER) in non-FIFO in-vehicle networks. It characterizes the adversarial framing scenarios that cause latency blow-ups, and proposes concrete ATS configuration guidelines—such as distributing ATS across all hops, tuning $cir$ and $cbs$, avoiding ATS behind FRER merges, and using the $mrt$ bound—to restore bounded latencies. Through OMNeT++/INET simulations and a realistic IVN case study, the authors demonstrate that these strategies can prevent unbounded delays in both synthetic networks and real-world-like topologies, albeit with trade-offs in reliability or resource usage. The work thereby supports the practical adoption of ATS in IVNs by outlining how to mitigate interactions with FRER and non-FIFO behavior, and points to future formal validation and cross-domain applications of the approach.

Abstract

Time-Sensitive Networking enhances Ethernet-based In-Vehicle Networks (IVNs) with real-time capabilities. Different traffic shaping algorithms have been proposed for time-critical communication, of which the Asynchronous Traffic Shaper (ATS) is an upcoming candidate. However, recent research has shown that ATS can introduce unbounded latencies when shaping traffic from non-FIFO systems. This impacts the applicability of ATS in IVNs, as these networks often use redundancy mechanisms, i.e. Frame Replication and Elimination for Reliability (FRER), that can cause non-FIFO behavior. In this paper, we approach the problem of accumulated delays from ATS by analyzing the scenarios that generate latency and by devising placement and configuration methods for ATS schedulers to prevent this behavior. We evaluate our approach in a simulation environment and show how it prevents conditions of unbounded delays. In an IVN simulation case study, we demonstrate the occurrence of unbounded latencies in a realistic scenario and validate the effectiveness of our solutions in avoiding them.

Figures (8)

• Figure 1: Assignment of frame eligibility times for two streams (black, red) scheduled by ATS in the same scheduler group.
• Figure 2: Adversarial frame generation sequence of three concurrent streams (blue, red, and orange) redrawn from tl-ncsjr-24.
• Figure 3: Switched network with adversarial frame generation (see Fig \ref{['fig:ats_unbounded_example']}). End systems are not shown. Each dotted line (blue, red, orange, and green) represents a different stream.
• Figure 4: Switch 1 and 2 of the example networks (see Fig. \ref{['fig:unbounded_networks']}). Dashed elements appear only in network A, while solid-line elements are present in networks A, B, and C. Colors (blue, red, and orange) represent the streams.
• Figure 5: Qualitative evolution of ATS tokens and eligibility times with two consecutively arriving frames of one stream.